Assessment of arsenic speciation and bioaccessibility in mine-impacted materials

Full text : Introduction - Operational and legacy gold (Au) mine sites exist throughout the world as a result of current and historical activities. These sites are numerous and are typically associated with elevated concentrations of arsenic (As) due to the association of both As and Au with sulphidic phases. Arsenic contamination at mine sites is a consequence of Au processing which results in three predominant contaminated materials; tailings, calcinated and slime materials. Tailings are typically associated with refuse material, i.e. materials left over after the process of separating the valuable fraction from the uneconomic fraction of an ore. Calcinated materials are waste products resulting from high temperature processes applied to ores that facilitates thermal decomposition, phase transition and the removal of volatile fractions. In contrast, slimes are fine materials, commonly associated with settling tanks and tailings ponds, produced during mechanical crushing that are separated from the valuable ore component. As a consequence of these physical and thermal processes, As speciation, As concentration and other physicochemical properties of the contaminated matrix will vary which may impact on the outcomes of As bioaccessibility testing. However, despite their different properties, 'mine wastes' are often reported under a generic descriptor which may under represent As bioaccessibility of individual mine-impacted materials and therefore impact on human health exposure assessment. In this study, mine-impacted materials were categorised and characterised to determine the influence of ore processing on As bioaccessibility. Materials and methods - Mine-impacted materials (n = 50) were collected from Victoria, Australia and categorized into three source materials based on physical appearance; tailings (n = 35), calcinated (n = 10) and slimes (n = 5). Total metal / metalloid concentrations in the < 2 mm and < 250 µm soil particle size fraction was determined using Inductively Coupled Plasma – Optical Emission Spectroscopy (ICP-OES) following acid digestion. Arsenic bioaccessibility in mine-impacted materials was determined using the <250 µm soil particle size fraction and the gastric phase of the Solubility Bioaccessibility Research Consortium (SBRC) assay. As speciation was determined using X-ray absorption near-edge structure (XANES). Results and discussion - The concentration of As in these materials varied over several orders of magnitude (30- 47,000 mg kg^-1), with median concentrations of 500, 10,800 and 1,500 mg kg^-1 for tailings, calcinated and slimes respectively. Other physicochemical properties varied between the mine-impacted materials notably total (2-110 g kg^-1) and amorphous Fe (0.18-14.2 g kg^-1) concentrations. When As bioaccessibility was assessed using the SBRC assay, As bioaccessibility was variable, with mean values of 30%, 49% and 82% for tailings, calcinated and slimes respectively. An analysis of variance (ANOVA) determined that As bioaccessibility was significantly different (P < 0.05) between source materials presumably due to differences in As speciation, in addition to the concentration of total and amorphous Fe which may influence As dissolution / precipitation under gastrointestinal conditions. XANES analysis identified arseniosiderite [Ca₂Fe³⁺₃(AsO₄)₃O₃•3(H₂O)], yukonite [Ca₇Fe³⁺₁₁(AsO₄)₉O₁₀•24.3(H₂O)], realgar [AsS], loellingite [Fe²⁺As₂] and mineral sorbed As^V species in mine-impacted materials with more crystalline forms prevalent in tailings and a greater proportion of mineral sorbed As^V present in calcinated materials. Conclusion - In recent years, research into mine-impacted materials has reported considerable variability in As bioaccessibility with a mean value of ~15%. Outcomes from this research highlight that variability in As bioaccessibility can be prescribed to As speciation and matrix physicochemical properties while categorizing samples into tailings, calcinated and slime materials can provide some notional indication of potential exposure. (author)